Ever since Open AI’s Chat GPT set a record for the number of users earlier this year, companies of all sizes have been trying to figure out how to harness the power of Generative AI. According to a global benchmark study by Lucidworks, 96% of AI decision-makers actively prioritize investments in generative AI, and 93% of companies plan to increase their AI spending in the next year.

Many companies start their generative AI journey with Open AI, often through a private cloud on Microsoft Azure. The Azure distribution provides companies with a private instance of the chatbot, ensuring that their proprietary data remains secure. Software provider Nerdio, for example, uses generative AI to create PowerShell scripts for its customers, convert code from one language to another, and develop a customized support chatbot.

While Open AI’s Chat GPT can handle many of these tasks, Nerdio’s custom support chatbot utilizes a different generative AI Model called text-embedding-ada-002. This model is specifically designed to work with embeddings, a type of database optimized for inputting data into large language models.

Open AI may have been the first player in the game, but it is no longer the only option. Companies are now considering other options such as Google’s Bard, Anthropics Claude, Databricks Dolly, Amazon’s Titan, or IBM’s Watson X. Open-source models like Llama 2 from Meta are also gaining popularity.

The availability and deployment of open-source models are becoming easier. Microsoft, for example, has announced its support for Llama 2 on its Azure cloud platform. AWS also offers support for several LLMs through its Amazon Bedrock service, including models from Anthropic, Stability AI, AI21 Labs, and Metas Llama 2.

Financial information company S&P Global Market Intelligence utilizes multiple models from different providers, including Microsoft, Google, Amazon, and open-source models from Hugging Face. Choosing the right LLM depends on the specific use case, and the company remains flexible to update or switch models as needed.

However, the focus should not solely be on finding the best model but on building flexible architectures. Building a solid architecture allows for easy integration of multiple LLMs and quick scalability.

Other platforms, such as Databricks, offer their own LLMs. Databricks’ open-source LLM, Dolly, is licensed for both research and commercial use and has gained popularity due to its compatibility with large datasets. Additionally, Snowflake’s Document AI provides a secure and private environment for running its LLM, allowing companies to protect customer data.

In conclusion, while Open AI remains a popular choice, companies are exploring alternative generative AI models from various players. The landscape is evolving rapidly, with major cloud providers and open-source projects contributing to the availability and accessibility of generative AI models. Building flexible architectures and prioritizing data security are crucial factors for successful implementation.

The post Using Generative AI: Open AI vs Other Players appeared first on TS2 SPACE.